September 2010

Vol. 2 No. 3

Mind the Gap:

Optimism vs. Pessimism in the Macro Investing
World

A
FTER SOME SPECTACULAR SUCCESSES
AT THE HEIGHT OF THE FINANCIAL
CRISIS, the macro hedge fund sector has
muddled along for much of 2009 and 2010. Many managers
have expressed pessimism about current macro opportunities,
believing that today’s fact set—highly correlated assets trading
in apparently range-bound fashion—presents a target-poor
climate in which to ply their trade. (Similarly negative views,
and for similar reasons, are widely seen in other investing
disciplines, including in long/short equities and credit.)
We’ve been hearing a common macro view that might be
stated along these lines: “We’re operating in a world in
which assets that used to move much more independently
are currently driven in lockstep by short-term news or by
whether investors are in ‘risk-on’ or ‘risk-off’ mode on a
given day. Predicting things like regulatory announcements
or changes in risk appetite among investors is very hard,
and that’s about all there is to bet on right now. Until
these elevated correlations come down, it’s going to be
tough to make a buck.”
On balance, we are very optimistic 1 about opportunities for
macro, and feel that this environment is in fact one of the
more compelling ones we’ve seen on the basis of its
offering broad (if not always deep) liquidity, coupled with a
large amount of pricing inefficiency. As a result, at a time
when we understand many macro managers are running
portfolios at reduced risk, we’ve been unusually active,
while remaining mindful as always that markets we
currently find attractive also embed a fair amount of risk.
What accounts for this difference of opinion between our
view of the risk-adjusted opportunity and the apparent
consensus among the broader macro community? We
outline our thoughts in this Market Insights.

1
For purposes of this Market Insights, “optimism“ and “pessimism” apply to the
opportunity for macro traders to spot and profit from inefficiencies, and not to
whether the global economy will prosper or deteriorate.
Mind the Gap
Different Macro Approaches,
Different Macro Opportunities
I
t’s our view that this macro opportunity perception gap
is partly of a structural nature. In particular, we believe
that our macro approach differs from that of a typical
macro manager in several respects: first, we scan for
trading opportunities across a wider range of asset classes;
second, we rely less on “predicting the future” in our
macro strategy, instead focusing on finding examples of the
market saying inconsistent things about the future; and
third, we tend to invest over longer horizons.
Looking Far Afield
Many macro practitioners focus largely (and in some cases
exclusively) on the sovereign fixed income and currency
markets, given those asset classes’ explicit linkage with
macroeconomic policy choices and data releases. We invest
extensively in these markets too, but we’re much more
active in commodities, equities, corporate credit, and
various forms of volatility than we think is typical of macro
managers.
Why is as much as 70 percent of our discretionary macro
strategy (in risk terms) often invested in products other than
fixed income or currencies? Among other things, we think
that by casting a wider net across asset classes, we end up
seeing a whole array of apparent mispricings across asset
groupings to an extent literally not possible for a macro team
(or, as is often the case, a relatively autonomous trading
“silo” at a larger macro outfit) focused narrowly on one asset
class. In fact, we think this kind of market segmentation is
directly responsible for some very interesting profit
opportunities.
For this discussion, we can isolate commodities as one asset
class that we think highly of as a hunting ground that is
often not a large focus of macro traders. The commodities
complex presents an unusually appealing blend of features:
it’s large, reasonably liquid, and has many subsectors
(energies, metals, ags, to name a few) with their own
idiosyncrasies. Many large commodities market participants
are motivated by considerations other than maximizing near-
term trading profits, and the alpha seekers focused on
commodities are often operating in extremely narrow niches
that may lever their expertise, but at the expense of
understanding broader relationships of their domain to other
MARKET INSIGHTS
assets. Right now, the commodities space accounts for
roughly 45 percent of our discretionary macro portfolio’s risk.
No Crystal Ball
Macro investing often involves a manager’s making some
forecast about the future, about global growth or upcoming
central bank moves, for example. People seem to agree that,
for various reasons—regulatory uncertainty, heightened
geopolitical risk, continued fallout from a financial crisis not
seen in most lifetimes—this is really hard to do right now.
In our case, we spend less time trying to make predictions
about the economic future and more time trying to find
different areas of the market that are implying different
predictions about the economic future. We then try to
understand some of the supply/demand factors that might
explain those apparently incompatible market statements.
The truth is that it’s really hard to add consistent macro value
by finding the market flat out “wrong.” Of course, it’s not
easy to invest in relative mispricings either, but we find that
seeking and analyzing these types of anomalies tends to
result in higher conviction trading views for us. Moreover,
given the firm’s experience in evaluating certain aspects of
these views (like correlations and crash risks), we think we
are unusually practiced in determining both which views are
worth expressing in our portfolio and in what size.
One prime example of the market saying interestingly
different things relates to everyone’s favorite subject, the
great inflation vs. deflation debate. Predictions about the
overall course global inflation will chart are, in our view,
difficult to make, given the dependency of that path on
political factors. But it’s interesting to see how inflationary
fears express themselves (or don’t) in such remarkably varied
ways, depending on whether one is looking at gold volatility,
inflation-linked bonds, interest-rate volatility, stocks in
inflation-sensitive industries, and so on. And it’s often clear
that differences in these asset classes’ apparent apprehension
of inflation are rooted in how actively inflation-fearful people
use those instruments to hedge their inflation risk. For
instance, gold options are an easy (if not necessarily very
efficient) way for investors to hedge against the possibility of
very high future inflation. On the other hand, picking out
inflation-sensitive versus non-inflation-sensitive stocks is
complicated and difficult to implement as a hedge. We think
the pricing of those assets reflects this in certain respects.
September 2010 | Vol. 2 No. 3 | Page 2 of 6
Mind the Gap

It’s Also About the Timing many individual assets away from prices that are
fundamentally justifiable. What are these technical forces?
There are naturally exceptions, but our macro strategy’s
There are a number of them, but they can be reduced in
sweet spot is for investments with expected horizons of
some sense to an exceptionally persistent “risk-on/risk-off”
from several weeks to several months, rather than the
tendency among investors since 2008. Investors are
much shorter horizon of a few days to a few weeks typical
understandably unsettled by the financial crisis and its
of some macro traders. We think this may account for
aftermath and are thus much more inclined to react abruptly
some of the difference between our and others’ views of
to new information. This can take a general form, as investors
the macro opportunity. Two instruments that a few years
quickly head to cash or hurriedly ramp up across a variety of
ago might have demonstrated meaningful—but far from
asset classes on the basis of short-term global economic
perfect—correlation might these days move in lock-step
pessimism or optimism. Alternatively, this risk-on/risk-off
fashion in response to a news item like a potential
mentality can take a narrower form, with investors reacting to
regulatory change. Someone operating on a one-week
news in a given sector by treating all assets in that sector the
horizon, perhaps enforced by rigid stop-loss rules, could
same, even if those assets actually have important
justifiably find this sort of phenomenon really frustrating,
distinguishing features relative to each other.
but we’ve seen that, over the intermediate to long term,
assets that may still exhibit unusually high (by historical Opportunity through Relative Price Drift
standards) correlations may nonetheless drift away from
Now let’s get back to our question of whether this correlation
each other in ways that we believe may allow for profitable
phenomenon is a positive or negative in terms of the macro
trading.
trading opportunity set. We believe these correlations signal a
lot of opportunity for certain macro investors, even after
Correlation: Friend or Foe?
factoring in elevated risk levels in the market.

W
e think the crucial difference between the
The macro trader focused on relatively short-term
optimistic and pessimistic macro worldview
opportunities, or looking only at a narrow asset class band,
hinges on the answer to this question: do the
or worried about getting stopped out of a trade due to
current correlation dynamics kill or create investment
hard-to-predict fluctuations in global risk tolerance, might
opportunities? We’ll give our answer in a moment, but
find this climate understandably exasperating. But we
let’s first try to account for the unusual correlation
believe that not only are there some very interesting trading
properties of today’s markets.
opportunities over a multi-month horizon, but that these
Imagine two assets, A and B, that have historically been sorts of heightened correlations are directly responsible for
somewhat correlated. Yet these days, the prices of A and B producing them.
move together in a much more synchronous fashion. What
The life cycle of these trades is straightforward. First, the
sort of event causes their short-term moves together?
prices of assets A and B move together in the short term
Maybe markets have broadly sold off on a given day due to
due to readily observable technical pressures that are
yet another sharp reversal in investor risk appetite, causing
otherwise contrary to the fundamentals of at least one of
A and B to trade down with them. Or maybe there’s
those instruments. Next, the prices of A and B end up
regulatory news causing all instruments remotely like A and
drifting gradually apart in meaningful ways as fundamentals
B to trade up or down in virtually identical fashion. The
slowly reassert themselves, even as the two instruments
point is that, in both examples, today’s market participants
remain, by historical standards, quite highly correlated on a
often trade A and B together a bit indiscriminately, with
day-to-day basis. Traders focused only on very short-term
insufficient reflection on the actual exposures of A and B to
effects remain understandably frustrated by that high
the global economy or the latest government initiative.
observed correlation, but, because we’re operating over a
Risk On/Risk Off longer horizon and are trying to diversify across a broad
number of asset classes and trade themes, we feel well
Put another way, in the current environment of shallow
positioned to attempt to take advantage of these
liquidity and heightened risk aversion, technical forces are
anomalies.
having an unusually strong effect on the market, driving

MARKET INSIGHTS September 2010 | Vol. 2 No. 3 | Page 3 of 6

Mind the Gap

Let’s make this more concrete with a couple of examples,
one relating to the relationship of sovereign bonds and
CDS, and one from the wonderful world of mortgages.
Swap Curve vs. Sovereign CDS
An example of an asset pairing that has recently exhibited
unusually high correlation is the relationship between credit
risk and the shape of the swap curve at the long end. If
one looks, for example, at the relationship between
sovereign CDS spreads and the euro-denominated 10s30s
spread, 2 it has recently been as negatively correlated as
-0.70, even though historically there has been no persistent
correlation. (See Figure 1 below.)
There are a few possible interpretations of this. A trader
could despair that, at least for a while, swap curve shape
and credit risk appear to be more or less the same thing,
and thus uninteresting as an alpha matter. For the reasons
mentioned earlier, we think this is not the right way to view
it. Instead, we see this strong relationship as a signal that
there may be an interesting and exploitable market
inefficiency here.
The long maturity end of the swap curve is typically
affected by many technical forces. For example, corporate
issuers of fixed-rate bonds often use swaps to hedge their
interest-rate exposure, which in practical terms means they
are “receiving fixed” and “paying floating” rates on swaps.
Pensions often use swaps similarly, as they receive fixed in
order to protect their plans from becoming underfunded if
interest rates fall.
Going the other direction, some governments use long-
dated swaps to pay fixed and receive floating in order to
protect themselves from a rise in rates, which would
otherwise increase the cost of their future debt issuance.
The impact of technicals associated with the sovereign
positions seems to have been especially dominant recently.
In particular, risk managers at banks that are exposed to
sovereign counterparties on swaps drive the correlation
between sovereign CDS spreads and the 10s30s spread
negative as they work to hedge their profits and risk. As
interest rates fall, for example, swap trades with sovereigns
become more profitable from the banks’ perspective.
Because many sovereigns do not post variation margin
under their swap agreements, the banks stand to lose their
unrealized profits if their sovereign counterparty were to
default. The banks thus need to go out and buy more
sovereign CDS as a hedge, driving the correlation negative.
The same result can be observed when sovereign CDS
spreads move. When sovereign CDS spreads rise, for
example, bank risk managers recognize that there is a
higher chance of a counterparty default. The banks
therefore have to put on more of the same receive-fixed
position with another counterparty as a precaution. Since
sovereigns tend to be more active at the thirty-year point
than at the ten-year point, this again drives the correlation
between sovereign CDS and the 10s30s spread negative.

10s30s Spread Versus Italy 5-Year CDS: Rolling 100-Day Correlation

0.4

0.2

0.0
Correlation

-0.2

-0.4

-0.6

-0.8
12/26/05 4/26/06 8/26/06 12/26/06 4/26/07 8/26/07 12/26/07 4/26/08 8/26/08 12/26/08 4/26/09 8/26/09 12/26/09 4/26/10 8/26/10

Figure 1 Sources: Bloomberg; The D. E. Shaw Group

2
“10s30s” is a shorthand notation referring to the spread
between 30-year swaps and 10-year swaps.

MARKET INSIGHTS September 2010 | Vol. 2 No. 3 | Page 4 of 6

Mind the Gap
So this is a case of an unusually high (in this case negative)
correlation being a sign of some interesting technicals at
play that we believe may present some trading
opportunities. If bank risk desks are motivated to transact
more by their short-term hedging program than by
considerations of fundamental worth, then this could lead
to some compelling trades for a market participant with an
eye towards a profit over the slightly longer term.
Fannie and Ginnie: Kissing Cousins
On February 10 of this year, government-sponsored
mortgage giants Fannie Mae and Freddie Mac announced
their intent to buy back from their mortgage-backed
securities pools a large portion of the underlying delinquent
mortgages. Under new accounting rules requiring
guaranteed securitized assets to be held on balance sheet,
the cost to Fannie and Freddie of loan guarantee payments
to the MBS holders had become greater than the cost to
the GSE of itself acquiring and holding the relevant
nonperforming underlying loans. The substantive effect of
this buyout was a massive prepayment to these GSEs’ MBS
holders. Since most of these MBSs were trading at a
significant premium to par, this effective prepayment was a
negative event for security holders. On the other hand,
Ginnie Mae has historically incorporated such buyouts into
its standard operating procedure, so in theory delinquency
buybacks would be priced into Ginnie Mae securities.
Figure 2 (below) shows the relationship of the Fannie Mae
6% MBS with the analogous Ginnie Mae MBS. It’s easy to
Fannie Mae and Ginnie Mae MBS: Price and Correlation
110
109
108
2/09/10:
Day before GSE's buyout
Price
107
106
105
1/19/10 2/19/10 3/19/10 4/19/10
Fannie Mae 6s (Price) Ginnie Mae 6s (Price)
Figure 2
MARKET INSIGHTS
see that the prices of both Fannie and Ginnie securities
reacted negatively to the Fannie and Freddie buyback
news, even though the news didn’t actually apply to Ginnie
Mae. Investors responded to the news of the Fannie and
Freddie buybacks by selling all mortgage securities trading
at a premium, even those with no actual fundamental
exposure to the event in question. Because of this investor
reaction and the resulting price action, one didn’t have to
take a “crystal ball” style directional view on whether the
mortgage market as a whole had overreacted or
underreacted to the announcement. Instead, one could
have focused on the relative overreaction of Ginnie Mae
prices based on a very clear technical force (panicky selling
of premium mortgages). Note that, even as correlations
remained extremely high between these two instruments
throughout the period in question, the two gradually (and we
believe correctly) drifted apart over a period of a few months
as the market implicitly recognized the two MBS’ different
fundamental exposures to buyout risk.
Looking for Gradual Drift is Not the Same as
Shorting Correlation
We should make one thing clear. While a reasonable
argument might be made that correlations are absolutely
too high, we don’t know whether and when the market will
come to that same conclusion. We are thus not asserting
that we want to short correlation. An express short
correlation bet is potentially quite dangerous, as it is less a
wager on the relationship of two instruments and more a
1.00
0.95
0.90
0.85
Correlation
0.80
0.75
0.70
0.65
5/19/10 6/19/10 7/19/10 8/19/10
Ginnie Mae 6s versus Fannie Mae 6s (Correlation)
Sources: Bloomberg; The D. E. Shaw Group
September 2010 | Vol. 2 No. 3 | Page 5 of 6
Mind the Gap

wager that the world has collectively decided to get off its Conclusion

current “risk-on,” “risk-off” seesaw. That is an altogether
different sort of prediction, and one that’s hard to make,
difficult to diversify, and apt to feel very uncomfortable
should the world enter another full-on financial crisis mode.
The potential alpha in trades like those we illustrate above
comes not from a secular fall in correlation, but instead
from a money manager’s ultimately being right that one
instrument moved too much relative to another instrument.
W
e are upbeat about current macro opportunities
given our focus on a wide range of assets, the
relatively longer horizon of our typical macro
investment, and our emphasis on understanding both the
fundamental and technical drivers of an asset’s pricing. The
world certainly presents its share of investment risks, but we
also believe it presents an unusual volume of interesting
inefficiencies for the investor willing and able to take a
longer view.

The views expressed in this commentary are solely those of the D. E. Shaw group as of the date of this commentary. The
views expressed in this commentary are subject to change without notice, and may not reflect the criteria employed by any
company in the D. E. Shaw group to evaluate investments or investment strategies. This commentary is provided to you for
informational purposes only. This commentary does not and is not intended to constitute investment advice, nor does it
constitute an offer to sell or provide or a solicitation of an offer to buy any security, investment product, or service. This
commentary does not take into account any particular investor’s investment objectives or tolerance for risk. The
information contained in this commentary is presented solely with respect to the date of the preparation of this
commentary, or as of such earlier date specified in this commentary, and may be changed or updated at any time without
notice to any of the recipients of this commentary (whether or not some other recipients receive changes or updates to the
information in this commentary).

No assurances can be made that any aims, assumptions, expectations, and/or objectives described in this commentary
would be realized or that the investment strategies described in this commentary would meet their objectives. None of the
companies in the D. E. Shaw group; nor their affiliates; nor any shareholders, partners, members, managers, directors,
principals, personnel, trustees, or agents of any of the foregoing shall be liable for any errors (to the fullest extent permitted
by law and in the absence of willful misconduct) in the information, beliefs, and/or opinions included in this commentary,
or for the consequences of relying on such information, beliefs, or opinions.

MARKET INSIGHTS September 2010 | Vol. 2 No. 3 | Page 6 of 6